Linking disease, demography and climate

1 08 2010

Last week I mentioned that a group of us from Australia were travelling to Chicago to work with Bob Lacy, Phil Miller, JP Pollak and Resit Akcakaya to make some pretty exciting developments in next-generation conservation ecology and management software. Also attending were Barry Brook, our postdocs: Damien Fordham, Thomas Prowse and Mike Watts, our colleague (and former postdoc) Clive McMahon, and a student of Phil’s, Michelle Verant. At the closing of the week-long workshop, I thought I’d share my thoughts on how it all went.

In a word, it was ‘productive’. It’s not often that you can spend 1 week locked in a tiny room with 10 other geeks and produce so many good and state-of-the-art models, but we certainly achieved more than we had anticipated.

Let me explain in brief why it’s so exciting. First, I must say that even the semi-quantitative among you should be ready for the appearance of ‘Meta-Model Manager (MMM)’ in the coming months. This clever piece of software was devised by JP, Bob and Phil to make disparate models ‘talk’ to each other during a population projection run. We had dabbled with MMM a little last year, but its value really came to light this week.

We used MMM to combine several different models that individually fail to capture the full behaviour of a population. Most of you will be familiar with the individual-based population viability (PVA) software Vortex that allows relatively easy PVA model building and is particular useful for predicting extinction risk of small populations. What you most likely don’t know exists is what Phil, Bob and JP call Outbreak – an epidemiological modelling software based on the classic susceptible-exposed-infectious-recovered framework. Outbreak is also an individual-based model that can talk directly to Vortex, but only through MMM.

I’ll use one example of our work to explain what happens. We are interested in understanding the dynamics of a disease like tuberculosis entering and spreading through the population of swamp buffalo in northern Australia (see previous post for all you need to know about swamp buffalo). We first set up an epidemiological model in Outbreak using some fairly well-quantified estimates of contact rate, transmission probability, latency period, and recovery of tuberculosis in cattle and other bovids on a daily time scale. We then built a annual-timescale demographic PVA in Vortex based on our measured estimates of mortality, fertility and sex ratio.

MMM then allows the two models to exchange information at the different time scales. The disease enters and spreads through the population following the daily dynamics, and then these outputs are read into Vortex at the end fo the year, modifying vital rates like survival and fertility according to each individual’s disease status. The population is projected one year forward, and the daily Outbreak dynamics are run again on the surviving individuals, and so on until the end of the projection interval.

Pretty bloody cool, no?

We then can test a couple of important management options should tuberculosis (or similar pathogen) ever re-enter the Australian population of buffalo. These include, what’s the probability of detecting the disease if present based on various sampling frequencies? How many animals must be culled to eradicate the disease? These questions will be answered for you in a few months once the paper is complete.

 We also worked on another example to combine the cohort-based modelling software RAMAS with Vortex and Outbreak via MMM. The example included the complex interaction of black-footed ferrets, prairie dogs and plague determining the extinction risk of the ferrets. For the first time ever, RAMAS and Vortex are now working together. We also plan to bring climate-change projections into the combination in the near future.

We anticipate punching out over the next 12 months a heap of manuscripts explaining all this for your benefit, and we sincerely hope that MMM will grow into standard platform for combining disparate, but related population models for all sorts of conservation and disease-management applications. Stay tuned and I’ll keep you informed.

We worked pretty hard this week and rewarded ourselves with a little trip Friday night to Buddy Guy’s Legends blues club in the heart of Chicago. Fantastic music; I’ve never seen Clive happier (he’s a massive Blues Brothers fan).

Many thanks especially to Bob, Phil and the Chicago Zoological Society for hosting the workshop. We’ll return the favour next year in Adelaide.

CJA Bradshaw

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to Yahoo BuzzAdd to Newsvine

Like This!



3 responses

26 01 2012
Saludos desde Mataelpino « BraveNewClimate

[…] over 200 bird species in the U.K. in relation to climate and land-use change, as well as developing a multi-species population viability analysis metapopulation model on the predator-prey-habitat interactions of the critically endangered Iberian lynx, rabbits, […]


2 11 2011
Where the sick buffalo roam «

[…] figured out a way of coupling the two via something they called MetaModel Manager (MMM – about which I blogged last year) which could allow the daily step of Outbreak to run effectively within the annual step of Vortex […]


7 08 2010
US Travel update, ‘Argonne West Diaries’ upcoming « BraveNewClimate

[…] First, The 6-day intensive workshop in Chicago was terrific, and Corey Bradshaw has done a great job of describing its outcomes. Rather than re-hash this, I’ll quote Corey: Linking disease, demography and climate […]


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: